Food Insecurity: A Trek Through Data

Author

Tori Edmunds, Parsa Keyvani, Hannah Kim, & Irene Tait

Introduction and Topic Overview

It’s tempting to begin this article with a gimmick. Something like, “Food, huh? We can all relate to that!” But the truth is that the way each of us relates to our food can look very different from the ways our fellow humans on this planet may relate to theirs. Yes, we eat for nourishment of our bodies, but for many other reasons as well: community, connection, entertainment, culture, health, or even simple survival. The decisions we make surrounding what to eat each day are simultaneously unique to us as individuals, and also deeply human and shared by all. At the same time, the factors that go into what options the world around us presents to us when it comes to making those decisions about our food vary wildly with the context an individual is situated in. This project is an attempt to understand more about how people around the world relate to food and, more specifically, to food in the context of other elements of their lives and situations. But before we dive straight into our data from the Food and Agriculture Organization of the United Nations, we’d like to zoom as far out as we can and talk a little bit about population.

Below is a line chart showing the population of each continental region over the last 50 years:

The first thing that jumps out to most viewers here is the distance between Asia’s population numbers and growth rate and the rest of the world’s. In 1970, Asia had more than twice the population of the second most populous continent, Europe. 50 years later, and Europe has been bumped down to the third place with very little growth at all - about 250 million people less than the new second place continent, Africa, and a whopping three billion people less than Asia. This was our first clue that we’d need to narrow down our focus when looking at the global metrics in our analysis below. Many of the data elements you’re about to see are per capita (aka per person) evaluations, or in some cases percentages of the total population of a country that fit into the given category. Some metrics go even further than that, being calculated on rolling three year averages, so that freezing a specific moment in time becomes quite difficult. How would such rapid population growth affect these more staid data processes? Is it possible to compare a per person evaluation of GDP (gross domestic product) or other such metrics when the population scales of the two entities being compared are so vastly different? We were not ready to completely narrow our focus to one specific region just yet, but we now had (and hope you do too) a better grasp of the scale of these metrics and what they may mean.

As we take a look at the UN’s data on the Suite of Food Security Indicators, we invite you to think more about your own relationship with hunger, health, security, and community. When you look at those lines in the graph above, do you immediately search out a particular region? Do you see yourself or those around you represented in this data, or simply aggregated away into a tiny fraction? This article will delve into a global discussion of how we can understand food security in the 21st century, what the effects of insecurity may be, and why it matters. While we do indeed all eat to survive, our relationship with food can be complicated and influence by our upbringing, community, and culture. When we talk about the prevalence of malnutrition, do you have a different instinctive response than when when we discuss the prevalence of obesity? Is food insecurity something you have experienced in your life, or is it a theoretical issue that happens Elsewhere? How about political violence? We hope you will keep these questions in mind as you explore the visualizations in the article below.

A small note before we begin: as some of our visualizations are served up by an external source, you will need a working internet connection to view the next two plots.

A Closer Look at the Food Supply: Why Does This Matter?

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.

Food Insecurity & Gender: Zooming In to Africa & Europe

For the remainder of the article we are going to focus on two continents - Africa, and Europe. We made this decision for several reasons. Primarily, we wanted to have two continents to compare and contrast with one another to cut down on the cognitive load we would place on our viewers. As discussed in the introduction, the rapid population growth and scale difference between Asia and any other continental region would make comparison difficult. We chose Africa and Europe out of the remaining continents for three reasons.

Africa is the most food insecure continent on the planet in terms of severity.*
After Asia, Africa and Europe are the most populous countries with similar populations in 2000 yet different rates of growth leading to differing population levels today.
Because of the ramifications of European colonization of Africa in the so-called Scramble for Africa from the mid 1800s to the early 20th century, African and European economies are linked together in specific and interesting ways.

We’ve looked closely at the relationship between region, GDP, and whether the average person in a country receives enough energy from their food to perform daily life. Now let’s narrow our focus and look at the percentage of individuals in a country who we know do not get enough to eat: that is, the percentage of adults who live in a household that the UN classifies as food severely food insecure. The following is a direct quotation from the description of this metric in the metadata section of our dataset:

*“The threshold to classify”severe” food insecurity corresponds to the severity associated with the item “having not eaten for an entire day” on the global FIES scale. In simpler terms, a household is classified as severely food insecure when at least one adult in the household has reported to have been exposed, at times during the year, to several of the most severe experiences described in the FIES questions, such as to have been forced to reduce the quantity of the food, to have skipped meals, having gone hungry, or having to go for a whole day without eating because of a lack of money or other resources. It is an indicator of lack of food access.” (Bolding unoriginal)

Let’s take a look at how severe food insecurity rates break down by country, when averaged over the last two decades and separated by gender:

As we can see from a glance, there is a massive difference in the prevalence of severe food insecurity between our two continents of interest, regardless of gender. Malawi and the Congo have the highest percentage of male adults living in severely food insecure households, with 17.3% and 16.8%, respectively; for female adults, we see the two highest rates in the Congo (17.1%) and Guinea (16.1%). This contrasts starkly with Europe, where we see the highest rates in Albania, with only 3.5% of male adults and 2.7% of female adults living in severe food insecurity, and Romania, with 1.6% of men and 1.3% of women living in households in the same category.

It is particularly remarkable that even for the most food insecure nations in each region, the difference between the food insecurity rate of men and women is quite small, within a percent. This was a surprise to us, and we wonder if it was a surprise to you as well. It appears that the prevalence of severe food insecurity in both regions is somewhat gender blind, affecting men and women at similar rates. Of course this does not take into account differences in population demographics; if a country’s population is more male than female, even a less than one percent difference in prevalence rates could mean thousands more men than women (or vice versa) living in such dire situations.

When discussing what dataset to use here, I confess that we as a group struggled a bit. There were conversations about using one level of severity over another, as well as whether or not to include children in this analysis. This is a heavy topic and not one taken lightly. Additionally it is rarely a good idea to compare one human’s suffering against another’s, or to say that the quantity of people suffering is the only metric that matters - that way lies madness. However, as a way to look at the state of the world, we stand by our decision to include troubling statistics like this one that highlight some of the inequities around us. We do not intend to minimize those individuals in Albania and Romania (or any country!) who make these impossible decisions about what or when or how much to eat, but merely to highlight the truly different scales that these issues exist on.

There is clearly a difference to the average person in these two regions in one’s ability to access sufficient foodstuffs. Next, we would like to take a look at two other metrics attempt to measure the impact that difference in access can have on the health of people in these regions.

Who Is Affected? Obesity and Malnutrition

Tracking Obesity and Malnutrition Throughout the 21st Century

Malnourishment and obesity represent two important and interconnected global health challenges. Malnourishment can be characterized as the inadequate intake of essential nutrients, whereas obesity is the excessive intake of nutrient-poor foods. Malnourishment affects millions worldwide, more often in low-income regions. Obesity affects millions across all ages, genders, and socioeconomic backgrounds. Addressing both health issues necessitates concerted efforts and global cooperation that prioritizes equality and sustainability.

From 2000 through 2016, Europe maintained a higher average obesity rate compared to Africa. Obesity continued to rise in both continents, although the growth was relatively small, reaching a maximum of 0.41% increase.

In 2016, both Europe and Africa recorded their highest average obesity rates of 22.74% and 11.78%, respectively. In 2000, both continents reported the lowest average obesity rates, at 16.83% in Europe and 6.78% in Africa. Notably, Europe’s lowest average obesity rate surpassed any recorded average obesity rate in Africa during this time period. These averages highlight a pronounced difference in obesity between Europe and Africa.

Between 2000 and 2016, Africa consistently exhibited a higher average malnourishment rate compared to Europe.

Africa’s average malnourishment rate peaked at 24.14% in 2000, while Europe reached its highest rate of 10.48% in 2004. Conversely, Africa achieved its lowest average malnourishment rate of 17.81% in 2011, whereas Europe hit its lowest average rate of 3.60% in 2016.

In Europe, the largest spike in malnourishment was from 2003 to 2004 at 2.86%. While we cannot be certain, this may be because of the outbreak of conflict occurring in Kosovo, and the ramifications of the Rose Revolution in Georgia. During the same time period, the average malnourishment rate actually decreased in Africa by 0.64%. Interestingly, Europe had a sharp decline in malnourishment from 2006 to 2010, totaling to 6.29%, perhaps suggesting a recovery from the earlier periods of conflict. Africa’s average malnourishment rate continually decreased from 2000 to 2011, amounting to 6.33%. These trends underscore a disparity in malnutrition prevalence between the two continents.

Visualizing Food Supply Disparity by Country

The overall average rate of malnourishment in Africa, when calculated over the 20-odd years since the turn of the century, stands at 20.29%, contrasting with Europe’s average rate of 2.88%. The overall average rate of obesity in Africa is 9.50%, while in Europe it is 19.51%. Based on these averages, it’s evident that obesity had a greater impact on Europe than Africa, while malnutrition had a more pronounced effect on Africa compared to Europe during this time period. These averages suggest an inverse relationship between the malnourishment rate and the obesity rate in these continents.

Furthermore, there were a handful of countries who impressively maintained a malnutrition rate and obesity rate of less than 10.0% These countries include Mali, Nigeria, Ghana, Mauritius, and Mauritania. Of these countries, the average obesity rates ranged from 6.01% to 9.57% and the average malnourishment rates ranged from 5.42% to 9.19%.

In Europe, Malta has the highest average obesity rate of 26.04%, while Libya leads in Africa with an average obesity rate of 27.79%. In contrast, Bosnia and Herzegovina have the lowest average obesity rate in Europe at 15.05%, whereas Ethiopia has the lowest average obesity rate in Africa at 3.04%. Remarkably, Europe’s lowest average obesity rate is approximately five times higher than Africa’s lowest average obesity rate.

In Europe, Albania has the highest average malnourishment rate at 6.04%, while Somalia has the highest average malnourishment rate in Africa at 66.39%. This indicates that Somalia’s average malnutrition rate exceeds that of Albania by more than tenfold during this period.

In Europe, 22 countries exhibited an average malnourishment rate of less than 2.50%. These countries had their rates denoted as “<2.5%” in the data. Consequently, we treated these rates as a constant of 2.5%. Those countries are Austria, Belgium, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Lithuania, Luxembourg, Malta, Norway, Poland, Portugal, Romania, Slovenia, Spain, Sweden, and Switzerland. In Africa, Tunisia holds the lowest average malnourishment rate of 3.65%.

Political Climate and Food Supply

We’ve spent some time looking at the relationship between economic disparity and food security, and some of the health impacts of that relationship. Now we’d like to add political stability into the mix.

The UN issues an annual Political Security Index to each country, representing the perception that there will be political violence in the country. It is important to not that this is not a simple process, and that this rating does not represent the likelihood of political violence, merely the perception of the global community that a country is more or less prone to it. This score tends to range between -2.5 (extremely high perception of the likelihood of political violence, aka “perceived politically unstable”) to 2.5 (extremely low perception of the likelihood of political violence, aka “perceived politically stable”), with a score of 0 acting as a separating boundary between perceived as politically unstable vs stable. For the ease of discussion, we’ll drop the repeated use of the word “perception” and let “political stability” represent the Political Security Index. Now, let’s see what the relationship is between economic power (as measured by GDP per capita, as we have seen above) and political stability since the turn of the century.

The first thing that may jump out at you with this plot is the disparity between Europe and Africa in both categories. The larger dots represent an average of each continent (not by population, but simply by country), and we can see how African countries on average have both low GDP per capita and low political stability; Europe, by contrast, on average has both higher GDP per capita and political stability. Europe has a much larger spread in terms of GDP per capita, with a clear trend: European countries with lower political stability tend to have smaller economies per citizen. Additionally, over time Europe has increased it’s GDP per capita, moving further to the right from where it started, with relatively little difference in political stability. This is in contrast to Africa; African countries on average have not seen a similar growth in GDP per capita since 2002, and have also dipped slightly in political stability. There is also stronger clustering in Africa in a way that we do not see with Europe; most of the more politically stable African nations do not see a marked increase in GDP per capita when compare to their political unstable neighbors. Despite these clear continental difference, one point of commonality can be found. We can see two major economic contracting points that spanned both continents, when almost every country’s GDP per capita saw a decrease from the year before: namely, the 2008-09 economic recession, and the COVID-19 pandemic in 2020.

Some larger jumps in political stability that we see when allowing time to progress on the plot can be put into historical context. Ukraine hovers around the 0 margin until 2014, when it dips into the deeply negative political index scores at below -2, one of the only European nations to reach such a low, and remains low until starting to creep back up slowly in 2019; this is likely due to the conflict with and subsequent invasion from Russia (which, interestingly enough, achieves it’s maximum high in 2018). In Africa, Libya shifts from being one of the wealthier African nations with a political stability on par with Europe in 2009, before immediately steeply dropping in political stability and fluctuating greatly in GDP per capita, with a political stability score well below -2 in 2014, where it remains today. We can attribute this in part to the tensions that arose in 2010 with the first stirrings of the Arab Spring and the outbreak in 2011 of the Libyan Civil War.

There are a myriad of factors that go into both political stability, GDP, and total population of a country, and it would be remiss of us to say that one is directly and linearly affected by the other. But it is clear from this graph that there is a strong relationship between the two. It may be that European countries are able to grow their GDP per capita because of their overall political stability, whereas many African nations struggle to do the same. Or perhaps the inverse is true, and lower GDP per capita tends to exacerbate the prevalence of political violence. There is also a proximity element that is important not to discount. If a country is politically unstable, that could lead to political unrest in their neighbors as issues spill across border; similarly,

We should also keep in mind that dividing GDP across an entire population via averaging loses any measurement of income or wealth inequality (inequalities which here in the United States have led to unrest). In that respect, a median income level per person may be a more telling metric to compare against politically stability, but sadly that data is not included in our analytical dataset. On a different note, it is also not possible to immediately pull out authoritarian or democratic governmental groups from this plot, and governmental structure affects both economic and political environments.

Now that we’ve spent some time discussing political stability and it’s relationship to a country’s income and population, let’s return to food. We’ll be revisiting one metric we saw in the beginning of this article, and adding a new one.

Above, we can watch in rolling three year averages the comparison of two food security metrics against political stability: protein supply (in grams per person per day) and dietary energy supply adequacy. Again we have two larger dots representing the averages taken across all countries during each period, and again we can see clear clustering, especially in the lefthand plot.

Looking at protein supply and political stability, we again see the same levels of average political stability as in the previous graph, with Europe averaging politically stable and Africa averaging politically unstable. We also see that the majority of countries in Europe have, for the average person in their society, more than 80 grams of protein a day, whereas only roughly a third of African countries do. Looking to the right at dietary energy supply adequacy, we again see a similar disparity, although with a larger intersection between the clusters and more intermingling. We also see that both continents have, when averaged together, sufficient dietary energy supply for their citizens. While several African nations dip below the 100% dietary energy adequacy rate (as indicated on the graph by a vertical grey line), many recover and return to above the 100% mark. By the end of the 20 year period depicted, both continents have improved their overall average energy supply from diet.

Let’s take a closer look at the bottom left quadrant of that dietary energy supply plot. If a country’s point is in that area of the plot, it indicates that during the three year period being considered, this country is perceived as politically unstable while the average person in this country does not receive sufficient energy from their diet, leading to undernourishment and malnutrition. Very few European countries make an appearance in this category: only Moldova, which begins the century in this category but improves its energy supply somewhere around 2009, to end the century comfortably around the 140% dietary energy adequacy mark despite remaining slightly politically unstable during the entire two decades. Moldova sits at an intersection of very complicated political elements, both internal and external, so it is difficult to pinpoint an exact historical moment that could account for the shift. Two possibilities are the end of ripple effects from either the 2006 wine exports crisis or, going even further back, the period of steep inflation and economic crisis from the mid 90s that ended in only 2001. Only three countries both start and end this 20 year period in this unstable/insufficient quadrant: the Central African Republic, Zimbabwe, and Liberia. Of those three countries, only the Central African Republic spends the entire two decades in this state. Sadly this is not surprising for those with even a passing familiarity with the Central African Republic’s history, as the country has been in a state of crisis with multiple civil wars since the turn of the century, with one still ongoing.

Final Thoughts and Future Work

Remember those questions from the beginning? Let’s revisit them.

We asked you about your relationship with food, and to think about how you bring any preconceptions to this topic. We also asked whether or not you saw yourself in the data presented. Now that you have spent some time with this data, your answers may have changed. Or they may have not! There is no wrong answer here. We do hope that you have been able to give a little bit more thought to your relationship with food and security. Odds are that if you’re reading this (hello professors), you’re a person who currently does not have to worry about where your next meal has come from. Maybe you never were; maybe that’s a part of your past you hope not to relive; or maybe you have friends, relatives, or community members who you know do, or did, have that worry in their lives. We as authors had some of our preconceived notions challenged when working with this data, and we hope reading this article and getting to play with the data like we did has given you a better understanding of this topic so central to the lives of millions of people around the globe.

There is certainly more to do with this data than we have done justice in our short article. The original dataset is massive. We sliced off only small pieces in order to focus in on specific metrics of food insecurity and health, based primarily on our interests and what we thought we would be able to meaningfully convey through visualizations. Some specific metrics we did not get to but were keen on exploring in future projects were rail density (perhaps representing how easily food can be moved across the country) and the relationship between cereal imports, dietary makeup, and health. There is other data out there to be combined with this dataset in new and interesting ways; we spent some time tracking the colonial histories behind African and European nations, but ultimately were not able to satisfactorily visualize that relationship with the tools we had time to master during this project.

Thank you for reading.